A new machine learning ensemble model for class imbalance problem of screening enhanced oil recovery methods

Pirizadeh, Mohsen; Alemohammad, Nafiseh; Manthouri, Mohammad; Pirizadeh, Meysam

doi:10.1016/j.petrol.2020.108214

Cited by 40 publications

(18 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For such projects, it is crucial to assess the potential and risks in advance; however, this is not easy to do. It is of practical interest to optimize, in advance, possible control parameters for waterflooding and other IOR/EOR projects [39,40]. Future research should focus on applying predictive ML models for more advanced types of IOR/EOR projects.…”

Section: Discussionmentioning

confidence: 99%

Machine Learning to Rate and Predict the Efficiency of Waterflooding for Oil Production

2022

View full text Add to dashboard Cite

Waterflooding is a widely used secondary oil recovery technique. The oil and gas industry uses a complex reservoir numerical simulation and reservoir engineering analysis to forecast production curves from waterflooding projects. The application of such standard methods at the stage of assessing the potential of a huge number of projects could be computationally inefficient and requires a lot of effort. This paper demonstrates the applicability of machine learning to rate the outcome of waterflooding applied to an oil reservoir. We also explore the relationship of project evaluations by operators at the final stages with several performance metrics for forecasting. Real data about several thousand waterflooding projects in Texas are used in the current study. We compare the ML models rankings of the waterflooding efficiency and the expert rankings. Linear regression models along with neural networks and gradient boosting on decision threes are considered. We show that machine learning models allow reducing computational complexity and can be useful for rating the reservoirs, with respect to the effectiveness of waterflooding.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine Learning to Rate and Predict the Efficiency of Waterflooding for Oil Production

2022

View full text Add to dashboard Cite

show abstract

“…Artificial intelligence (AI) is a discipline that employs complex algorithms and networking tools to solve multidimensional problems by imitating human brainpower allowing machines to perform computational tasks (Yousef et al, 2020). Machine learning is an area of AI that deals with algorithm design and development to enable computers to use empirical data and learn behaviors or patterns (Zhang et al, 2020;Pirizadeh et al, 2021). The machine learning models find relations between inputs and outputs.…”

Section: Machine Learning In Petroleum Industry and Eormentioning

confidence: 99%

Enhanced oil recovery by nanoparticles ﬂooding: From numerical modeling improvement to machine learning prediction

Alwated

El-Amin

2021

Adv. Geo-Energy Res.

View full text Add to dashboard Cite

Nowadays, enhanced oil recovery using nanoparticles is considered an innovative approach to increase oil production. This paper focuses on predicting nanoparticles transport in porous media using machine learning techniques including random forest, gradient boosting regression, decision tree, and artificial neural networks. Due to the lack of data on nanoparticles transport in porous media, this work generates artificial datasets using a numerical model that are validated against experimental data from the literature. Six experiments with different nanoparticles types with various physical features are selected to validate the numerical model. Therefore, the researchers produce six datasets from the experiments and create an additional dataset by combining all other datasets. Also, data preprocessing, correlation, and features importance methods are investigated using the Scikit-learn library. Moreover, hyperparameters tuning are optimized using the GridSearchCV algorithm. The performance of predictive models is evaluated using the mean absolute error, the R-squared correlation, the mean squared error, and the root mean squared error. The results show that the decision tree model has the best performance and highest accuracy in one of the datasets. On the other hand, the random forest model has the lowest root mean squared error and highest R-squared values in the rest of the datasets, including the combined dataset.

show abstract

“…However, the main limitation of this type of oils is their high viscosity and density, which challenge and complicate their exploitation and production. − Therefore, research on enhanced heavy oil recovery (EOR) methods is essential to increase the rate of heavy oil production on one hand and on the other hand to improve the quality of the obtained oil as well . Until now, water injection, chemical injection, gas, and thermal injection have been considered the main proposed methods in the field of EOR. ,,,− However, thermal methods are widely believed to be the most important enhanced oil recovery methods in terms of the produced oil quality and economic aspects. According to the literature, in situ combustion (ISC) as a subset of the thermal injection method has been always considered as an effective and appropriate technology, especially when it comes to heavy oil reservoir exploitation. − Basically, this technology consists mainly of air injection into the reservoir under high pressure to provide the necessary conditions for initiating oxidation reactions and promoting a stable combustion flame front for a better oil recovery factor.…”

Section: Introductionmentioning

confidence: 99%

Effect of Ligand Structure on the Kinetics of Heavy Oil Oxidation: Toward Biobased Oil-Soluble Catalytic Systems for Enhanced Oil Recovery

Farhadian

Khelkhal

Tajik

et al. 2021

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

In recent years, protection of the environment from activities associated with enhanced oil recovery has been considered a crucial priority for decision-makers in the international community. The in situ combustion process as a promising thermal enhanced oil recovery method has been attracting considerable interest in terms of improving oil production and environmental protection. However, this technique is not yet well studied. This paper outlines a new approach to improve the process of heavy oil oxidation by designing new biobased oil-soluble catalysts that are able to maintain and stabilize the combustion flame front of the in situ combustion process. A comprehensive theoretical and experimental study including thermal analysis (thermogravimetry/differential scanning calorimetry, TG/DSC) and quantum calculations was used to shed light on the effect of the ligand structure in the oil-soluble catalytic system on the heavy oil oxidation process. The obtained accurate results proved that metal interaction with the designed ligands increased, which led to a decrease in the energy of activation and an increase in the heavy oil oxidation reaction rate. Besides, the obtained DSC curves showed one peak in the presence of Cu and biobased ligands, contrary to the curves obtained for heavy oil oxidation reported with other ligands and metals. In other words, the obtained catalysts merged low-temperature-and high-temperature oxidation regions into one region. These findings reveal that the structure of ligands can significantly affect their interaction with metals in oil-soluble catalysts, and therefore the efficiency of catalysts is dramatically improved. We believe that our work could be usefully employed for further studies to clarify the mechanism of the in situ combustion behavior of heavy oil for a better impact on the environment.

show abstract

A new machine learning ensemble model for class imbalance problem of screening enhanced oil recovery methods

Cited by 40 publications

References 52 publications

Machine Learning to Rate and Predict the Efficiency of Waterflooding for Oil Production

Machine Learning to Rate and Predict the Efficiency of Waterflooding for Oil Production

Enhanced oil recovery by nanoparticles ﬂooding: From numerical modeling improvement to machine learning prediction

Effect of Ligand Structure on the Kinetics of Heavy Oil Oxidation: Toward Biobased Oil-Soluble Catalytic Systems for Enhanced Oil Recovery

Contact Info

Product

Resources

About